Wednesday, 19 April 2017

The Basics of Forex Algorithmic Trading

Nearly thirty years ago, the foreign exchange market (Forex) was characterised by trades conducted via telephone, institutional investors, opaque price information, a clear distinction between inter dealer trading and dealer-customer trading and low market concentration. Today, technological advancements have transformed the market. Trades are primarily made via computers, allowing retail traders to enter the market, real-time streaming prices have led to greater transparency and the distinction between dealers and their most sophisticated customers has largely disappeared.

One particularly significant change is the introduction of algorithmic trading, which, while making significant improvements to the functioning of Forex trading, also poses a number of risks. By looking at the basics of the Forex market and algorithmic trading, we will identify some advantages algorithmic trading has brought to currency trading while also pointing out some of the risks.

Forex is the virtual place in which currency pairs are traded in varying volumes according to quoted prices whereby a base currency is given a price in terms of a quote currency. Operating 24 hours a day, five days a week, Forex is considered to be world's largest and most liquid financial market. Per the Bank for International Settlements (BIS) the daily global average volume of trading in April 2013 was $2.0 trillion. The bulk of this trading is done for U.S. dollars, euros and Japanese yen and involves a range of players, including private banks, central banks, pension funds, institutional investors, large corporations, financial companies and individual retail traders.

Although speculative trading may be the main motivation for certain investors, the primary reason for the Forex market’s existence is that people need to trade currencies in order to buy foreign goods and services. Activity in the Forex market affects real exchange rates and can therefore profoundly affect the output, employment, inflation and capital flows of any particular nation. For this reason, policymakers, the public and the media all have a vested interest in what goes on in the Forex market.

Basics of Algorithmic Trading

An algorithm is essentially a set of specific rules designed to complete a clearly defined task. In financial market trading, computers carry out user-defined algorithms characterized by a set of rules consisting of parameters such as timing, price or quantity that structure the trades that will be made.

Four Basic Types of Algorithmic Trading

There are four basic types of algorithmic trading within financial markets: statistical, auto-hedging, algorithmic execution strategies and direct market access. Statistical refers to an algorithmic strategy that looks for profitable trading opportunities based on the statistical analysis of historical time series data. Auto-hedging is a strategy that generates rules to reduce a trader’s exposure to risk. The goal of algorithmic execution strategies is to execute a predefined objective; such as reduce market impact or execute a trade quickly. Finally, direct market access describes the optimal speeds and lower costs at which algorithmic traders can access and connect to multiple trading platforms.

One of the subcategories of algorithmic trading is high frequency trading, which is characterised by the extremely high frequency of trade order executions. High-speed trading can give significant advantages to traders by giving them the ability to make trades within milliseconds of incremental price changes, but it may also carry certain risks.

Much of the growth in algorithmic trading in Forex markets over the past years has been due to algorithms automating certain processes and reducing the hours needed to conduct foreign exchange transactions. The efficiency created by automation leads to lower costs in carrying out these processes. One such process is the execution of trade orders. Automating the trading process with an algorithm that trades based on predetermined criteria, such as executing orders over a specified period of time or at a specific price, is significantly more efficient than manual execution by humans.

Banks have also taken advantage of algorithms that are programmed to update prices of currency pairs on electronic trading platforms. These algorithms increase the speed at which banks can quote market prices while simultaneously reducing the number of manual working hours it takes to quote prices.

Some banks program algorithms to reduce their exposure to risk. The algorithms may be used to sell a particular currency to match a customer’s trade in which the bank bought the equivalent amount in order to maintain a constant quantity of that particular currency. This allows the bank to maintain a pre-specified level of risk exposure for holding that currency.

These processes have been made significantly more efficient by algorithms, leading to lower transaction costs. Yet, these are not the only factors that have been driving the growth in Forex algorithmic trading. Algorithms have increasingly been used for speculative trading as the combination of high frequency and the algorithm’s ability to interpret data and execute orders has allowed traders to exploit arbitrage opportunities arising from small price deviations between currency pairs.

All of these advantages have led to the increased use of algorithms in the Forex market, but let’s look at some of the risks that accompany algorithmic trading.

Risks Involved in Algorithmic Forex Trading

Although algorithmic trading has made many improvements, there are some downsides that could threaten the stability and liquidity of the Forex market. One such downside relates to imbalances in trading power of market participants. Some participants have the means to acquire sophisticated technology that allows them to obtain information and execute orders at a much quicker speed than others. This imbalance between the haves and have-nots in terms of the most sophisticated algorithmic technology could lead to fragmentation within the market that may lead to liquidity shortages over time.

Furthermore, while there are fundamental differences between stock markets and the Forex market, there are some who fear that the high frequency trading that exacerbated the stock market flash crash on May 6, 2010 could similarly affect the Forex market. As algorithms are programmed for specific market scenarios, they may not respond quickly enough if the market were to drastically change. In order to avoid this scenario markets may need to be monitored and algorithmic trading suspended during market turbulence. However, in such extreme scenarios, a simultaneous suspension of algorithmic trading by numerous market participants could result in high volatility and a drastic reduction in market liquidity.

The Bottom Line

Although algorithmic trading has been able to increase efficiency, therefore reducing the costs of trading currencies, it has also come with some added risks. For currencies to function properly, they must be somewhat stable stores of value and be highly liquid. Thus, it is important that the Forex market remain liquid with low price volatility.

As with all areas of life, new technology introduces many benefits, but it also comes with new risks. The challenge for the future of algorithmic Forex trading will be how to institute changes that maximize the benefits while reducing the risks.

Reference: Matthew Johnston

Want to stay updated? See our Free Weekly Reports.

1 comment:

  1. How etf trading works?
    Automated Trading Strategies
    An etf trading, known as exchange traded fund is a marketable safety that tracks an index or basket of index.